3035 lines
81 KiB
JSON
3035 lines
81 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.0876731544800982,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742551231384278,
|
|
"epoch": 0.000292243848266994,
|
|
"grad_norm": 4.9375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7723,
|
|
"mean_token_accuracy": 9.259259095415474e-05,
|
|
"num_tokens": 11114.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742591094970702,
|
|
"epoch": 0.000584487696533988,
|
|
"grad_norm": 5.4375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7466,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 20935.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742585849761962,
|
|
"epoch": 0.000876731544800982,
|
|
"grad_norm": 5.09375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7483,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 31930.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742614936828613,
|
|
"epoch": 0.001168975393067976,
|
|
"grad_norm": 5.125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6676,
|
|
"mean_token_accuracy": 0.0005568421445786953,
|
|
"num_tokens": 41672.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742519950866699,
|
|
"epoch": 0.00146121924133497,
|
|
"grad_norm": 4.9375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.5834,
|
|
"mean_token_accuracy": 0.013186294632032514,
|
|
"num_tokens": 52510.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742103958129883,
|
|
"epoch": 0.001753463089601964,
|
|
"grad_norm": 4.3125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4654,
|
|
"mean_token_accuracy": 0.04129958841949701,
|
|
"num_tokens": 63214.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.740110874176025,
|
|
"epoch": 0.002045706937868958,
|
|
"grad_norm": 3.203125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.305,
|
|
"mean_token_accuracy": 0.05048500820994377,
|
|
"num_tokens": 73942.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.733905601501466,
|
|
"epoch": 0.002337950786135952,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1895,
|
|
"mean_token_accuracy": 0.051587145403027536,
|
|
"num_tokens": 84526.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.72324333190918,
|
|
"epoch": 0.0026301946344029457,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.1139,
|
|
"mean_token_accuracy": 0.04853160977363587,
|
|
"num_tokens": 95178.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.714754772186279,
|
|
"epoch": 0.00292243848266994,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.0146,
|
|
"mean_token_accuracy": 0.05312369242310524,
|
|
"num_tokens": 106653.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.708310794830322,
|
|
"epoch": 0.0032146823309369336,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.9113,
|
|
"mean_token_accuracy": 0.050825309753417966,
|
|
"num_tokens": 115969.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.704185581207275,
|
|
"epoch": 0.003506926179203928,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8939,
|
|
"mean_token_accuracy": 0.048680877685546874,
|
|
"num_tokens": 125826.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.701387882232666,
|
|
"epoch": 0.0037991700274709215,
|
|
"grad_norm": 1.890625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.872,
|
|
"mean_token_accuracy": 0.04808175079524517,
|
|
"num_tokens": 136677.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.697122478485108,
|
|
"epoch": 0.004091413875737916,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.8236,
|
|
"mean_token_accuracy": 0.04725121222436428,
|
|
"num_tokens": 147535.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.691372871398926,
|
|
"epoch": 0.00438365772400491,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.722,
|
|
"mean_token_accuracy": 0.04985377714037895,
|
|
"num_tokens": 157524.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.681602478027344,
|
|
"epoch": 0.004675901572271904,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6896,
|
|
"mean_token_accuracy": 0.05114552080631256,
|
|
"num_tokens": 168344.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.67192096710205,
|
|
"epoch": 0.004968145420538897,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.6618,
|
|
"mean_token_accuracy": 0.04409446604549885,
|
|
"num_tokens": 178976.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.664519691467286,
|
|
"epoch": 0.0052603892688058915,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.5639,
|
|
"mean_token_accuracy": 0.05372482612729072,
|
|
"num_tokens": 190567.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.647130489349365,
|
|
"epoch": 0.005552633117072886,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.4984,
|
|
"mean_token_accuracy": 0.05118483640253544,
|
|
"num_tokens": 201587.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.62646837234497,
|
|
"epoch": 0.00584487696533988,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.4193,
|
|
"mean_token_accuracy": 0.04868996068835259,
|
|
"num_tokens": 211913.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.610049438476562,
|
|
"epoch": 0.006137120813606874,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.3409,
|
|
"mean_token_accuracy": 0.05402343980967998,
|
|
"num_tokens": 222247.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.584276962280274,
|
|
"epoch": 0.006429364661873867,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.2142,
|
|
"mean_token_accuracy": 0.06452971324324608,
|
|
"num_tokens": 232361.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.533616924285889,
|
|
"epoch": 0.006721608510140861,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.1281,
|
|
"mean_token_accuracy": 0.0630628488957882,
|
|
"num_tokens": 242885.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.462300109863282,
|
|
"epoch": 0.007013852358407856,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.0436,
|
|
"mean_token_accuracy": 0.060920076444745064,
|
|
"num_tokens": 254482.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.408651351928711,
|
|
"epoch": 0.00730609620667485,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.93,
|
|
"mean_token_accuracy": 0.059300025925040245,
|
|
"num_tokens": 265744.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.367331790924073,
|
|
"epoch": 0.007598340054941843,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.8327,
|
|
"mean_token_accuracy": 0.058924198523163794,
|
|
"num_tokens": 276924.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.272025775909423,
|
|
"epoch": 0.007890583903208837,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.7304,
|
|
"mean_token_accuracy": 0.0622296292334795,
|
|
"num_tokens": 287779.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.183625984191895,
|
|
"epoch": 0.008182827751475831,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.6335,
|
|
"mean_token_accuracy": 0.06356315091252326,
|
|
"num_tokens": 298807.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.097103500366211,
|
|
"epoch": 0.008475071599742826,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.5907,
|
|
"mean_token_accuracy": 0.0559815414249897,
|
|
"num_tokens": 309760.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.004411506652833,
|
|
"epoch": 0.00876731544800982,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.397,
|
|
"mean_token_accuracy": 0.06126284077763557,
|
|
"num_tokens": 319750.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.832055950164795,
|
|
"epoch": 0.009059559296276814,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.2924,
|
|
"mean_token_accuracy": 0.0603407833725214,
|
|
"num_tokens": 329931.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.66302547454834,
|
|
"epoch": 0.009351803144543808,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.2478,
|
|
"mean_token_accuracy": 0.06180734224617481,
|
|
"num_tokens": 340457.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.49033498764038,
|
|
"epoch": 0.009644046992810802,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.1199,
|
|
"mean_token_accuracy": 0.06424234174191952,
|
|
"num_tokens": 352450.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.311860942840577,
|
|
"epoch": 0.009936290841077795,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.1159,
|
|
"mean_token_accuracy": 0.06317566409707069,
|
|
"num_tokens": 363204.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.153958892822265,
|
|
"epoch": 0.010228534689344789,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.0882,
|
|
"mean_token_accuracy": 0.062278729677200315,
|
|
"num_tokens": 373577.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.929795932769775,
|
|
"epoch": 0.010520778537611783,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.9663,
|
|
"mean_token_accuracy": 0.07208616435527801,
|
|
"num_tokens": 384267.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.768066787719727,
|
|
"epoch": 0.010813022385878777,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.9324,
|
|
"mean_token_accuracy": 0.06511378176510334,
|
|
"num_tokens": 394767.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.666393280029297,
|
|
"epoch": 0.011105266234145771,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.9172,
|
|
"mean_token_accuracy": 0.07024285569787025,
|
|
"num_tokens": 405807.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.585619068145752,
|
|
"epoch": 0.011397510082412765,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.869,
|
|
"mean_token_accuracy": 0.0643521461635828,
|
|
"num_tokens": 416381.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.565132427215577,
|
|
"epoch": 0.01168975393067976,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.9632,
|
|
"mean_token_accuracy": 0.06446416340768338,
|
|
"num_tokens": 427262.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.418638515472413,
|
|
"epoch": 0.011981997778946754,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.8962,
|
|
"mean_token_accuracy": 0.06822493113577366,
|
|
"num_tokens": 438504.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.47780933380127,
|
|
"epoch": 0.012274241627213748,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.8873,
|
|
"mean_token_accuracy": 0.06689661256968975,
|
|
"num_tokens": 449923.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.430081558227538,
|
|
"epoch": 0.01256648547548074,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.8463,
|
|
"mean_token_accuracy": 0.0716497365385294,
|
|
"num_tokens": 460736.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.367777156829835,
|
|
"epoch": 0.012858729323747735,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.9365,
|
|
"mean_token_accuracy": 0.0605232123285532,
|
|
"num_tokens": 471513.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.417449378967286,
|
|
"epoch": 0.013150973172014729,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.7879,
|
|
"mean_token_accuracy": 0.07277968265116215,
|
|
"num_tokens": 482115.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.359326553344726,
|
|
"epoch": 0.013443217020281723,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.8079,
|
|
"mean_token_accuracy": 0.07002997919917106,
|
|
"num_tokens": 493273.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.321810531616212,
|
|
"epoch": 0.013735460868548717,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.8188,
|
|
"mean_token_accuracy": 0.07201721332967281,
|
|
"num_tokens": 503939.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.311158466339112,
|
|
"epoch": 0.014027704716815711,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.7369,
|
|
"mean_token_accuracy": 0.06772977598011494,
|
|
"num_tokens": 515215.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.281570434570312,
|
|
"epoch": 0.014319948565082705,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.8223,
|
|
"mean_token_accuracy": 0.07254013754427432,
|
|
"num_tokens": 526485.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.277181148529053,
|
|
"epoch": 0.0146121924133497,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.7672,
|
|
"mean_token_accuracy": 0.06958542391657829,
|
|
"num_tokens": 536735.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.277847576141358,
|
|
"epoch": 0.014904436261616694,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.8541,
|
|
"mean_token_accuracy": 0.06499786637723445,
|
|
"num_tokens": 547326.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.180327033996582,
|
|
"epoch": 0.015196680109883686,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.7502,
|
|
"mean_token_accuracy": 0.06869623251259327,
|
|
"num_tokens": 558275.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.257558250427246,
|
|
"epoch": 0.01548892395815068,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.7096,
|
|
"mean_token_accuracy": 0.07516702339053154,
|
|
"num_tokens": 569711.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.150344276428223,
|
|
"epoch": 0.015781167806417674,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.6881,
|
|
"mean_token_accuracy": 0.07785917520523071,
|
|
"num_tokens": 579602.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.24105463027954,
|
|
"epoch": 0.01607341165468467,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.7286,
|
|
"mean_token_accuracy": 0.07401030212640762,
|
|
"num_tokens": 590899.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.173699569702148,
|
|
"epoch": 0.016365655502951663,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.726,
|
|
"mean_token_accuracy": 0.06962668001651764,
|
|
"num_tokens": 602080.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.16687994003296,
|
|
"epoch": 0.016657899351218655,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.7605,
|
|
"mean_token_accuracy": 0.07070777006447315,
|
|
"num_tokens": 613692.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.145956993103027,
|
|
"epoch": 0.01695014319948565,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.7305,
|
|
"mean_token_accuracy": 0.07069781571626663,
|
|
"num_tokens": 625019.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.123618793487548,
|
|
"epoch": 0.017242387047752643,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.6456,
|
|
"mean_token_accuracy": 0.07288045659661294,
|
|
"num_tokens": 636453.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.108728456497193,
|
|
"epoch": 0.01753463089601964,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.6918,
|
|
"mean_token_accuracy": 0.07240097671747207,
|
|
"num_tokens": 647272.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.040448427200317,
|
|
"epoch": 0.017826874744286632,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.701,
|
|
"mean_token_accuracy": 0.07080473527312278,
|
|
"num_tokens": 657724.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.166200351715087,
|
|
"epoch": 0.018119118592553628,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.7053,
|
|
"mean_token_accuracy": 0.07137923240661621,
|
|
"num_tokens": 669183.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.084780406951904,
|
|
"epoch": 0.01841136244082062,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.6679,
|
|
"mean_token_accuracy": 0.06901021562516689,
|
|
"num_tokens": 680439.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.157064056396484,
|
|
"epoch": 0.018703606289087616,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.712,
|
|
"mean_token_accuracy": 0.0706634670495987,
|
|
"num_tokens": 690789.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.046548128128052,
|
|
"epoch": 0.01899585013735461,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.6831,
|
|
"mean_token_accuracy": 0.07187991701066494,
|
|
"num_tokens": 701818.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.062455224990845,
|
|
"epoch": 0.019288093985621604,
|
|
"grad_norm": 0.71484375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.6316,
|
|
"mean_token_accuracy": 0.07437365353107453,
|
|
"num_tokens": 713344.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.084509325027465,
|
|
"epoch": 0.019580337833888597,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.6384,
|
|
"mean_token_accuracy": 0.07300028279423713,
|
|
"num_tokens": 723279.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.018547964096069,
|
|
"epoch": 0.01987258168215559,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.625,
|
|
"mean_token_accuracy": 0.07260205298662185,
|
|
"num_tokens": 733875.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.072830390930175,
|
|
"epoch": 0.020164825530422585,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.6669,
|
|
"mean_token_accuracy": 0.0713542215526104,
|
|
"num_tokens": 745449.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.05644793510437,
|
|
"epoch": 0.020457069378689578,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.6741,
|
|
"mean_token_accuracy": 0.0740118071436882,
|
|
"num_tokens": 757570.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.017089939117431,
|
|
"epoch": 0.020749313226956573,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.5737,
|
|
"mean_token_accuracy": 0.08025477081537247,
|
|
"num_tokens": 767601.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.00361680984497,
|
|
"epoch": 0.021041557075223566,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.6323,
|
|
"mean_token_accuracy": 0.0752041395753622,
|
|
"num_tokens": 777852.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.018228721618652,
|
|
"epoch": 0.021333800923490562,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.5859,
|
|
"mean_token_accuracy": 0.0751498755067587,
|
|
"num_tokens": 788339.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.970228672027588,
|
|
"epoch": 0.021626044771757554,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.6588,
|
|
"mean_token_accuracy": 0.07353257052600384,
|
|
"num_tokens": 799791.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.015653133392334,
|
|
"epoch": 0.02191828862002455,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.5746,
|
|
"mean_token_accuracy": 0.07374845892190933,
|
|
"num_tokens": 810141.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.040093088150025,
|
|
"epoch": 0.022210532468291543,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.6182,
|
|
"mean_token_accuracy": 0.07429262697696686,
|
|
"num_tokens": 820929.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.9526266098022464,
|
|
"epoch": 0.022502776316558535,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.5519,
|
|
"mean_token_accuracy": 0.07891798727214336,
|
|
"num_tokens": 831340.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.935062026977539,
|
|
"epoch": 0.02279502016482553,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.5636,
|
|
"mean_token_accuracy": 0.0735638827085495,
|
|
"num_tokens": 841846.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.029356002807617,
|
|
"epoch": 0.023087264013092523,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.5434,
|
|
"mean_token_accuracy": 0.07343484573066235,
|
|
"num_tokens": 852089.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.838059997558593,
|
|
"epoch": 0.02337950786135952,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.5241,
|
|
"mean_token_accuracy": 0.07688843682408333,
|
|
"num_tokens": 862568.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.034818124771117,
|
|
"epoch": 0.02367175170962651,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.5905,
|
|
"mean_token_accuracy": 0.07528307065367698,
|
|
"num_tokens": 874676.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.867450904846192,
|
|
"epoch": 0.023963995557893508,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.5115,
|
|
"mean_token_accuracy": 0.08100188933312893,
|
|
"num_tokens": 886338.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.946144676208496,
|
|
"epoch": 0.0242562394061605,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.5287,
|
|
"mean_token_accuracy": 0.07384060397744178,
|
|
"num_tokens": 896671.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.923208522796631,
|
|
"epoch": 0.024548483254427496,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.5025,
|
|
"mean_token_accuracy": 0.07231347896158695,
|
|
"num_tokens": 908265.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.851961994171143,
|
|
"epoch": 0.02484072710269449,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.4692,
|
|
"mean_token_accuracy": 0.07885380312800408,
|
|
"num_tokens": 919024.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.894201850891113,
|
|
"epoch": 0.02513297095096148,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.4639,
|
|
"mean_token_accuracy": 0.08095669150352477,
|
|
"num_tokens": 930837.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.845162343978882,
|
|
"epoch": 0.025425214799228477,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.4312,
|
|
"mean_token_accuracy": 0.076503673940897,
|
|
"num_tokens": 941374.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.838292598724365,
|
|
"epoch": 0.02571745864749547,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.3796,
|
|
"mean_token_accuracy": 0.0836145281791687,
|
|
"num_tokens": 953273.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.836203670501709,
|
|
"epoch": 0.026009702495762465,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.3659,
|
|
"mean_token_accuracy": 0.09230264946818352,
|
|
"num_tokens": 964882.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.860473203659057,
|
|
"epoch": 0.026301946344029457,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.5208,
|
|
"mean_token_accuracy": 0.08153782263398171,
|
|
"num_tokens": 974732.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.774071216583252,
|
|
"epoch": 0.026594190192296453,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.4513,
|
|
"mean_token_accuracy": 0.07568450495600701,
|
|
"num_tokens": 985709.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.999471426010132,
|
|
"epoch": 0.026886434040563446,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.4197,
|
|
"mean_token_accuracy": 0.07974797300994396,
|
|
"num_tokens": 998177.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.711365842819214,
|
|
"epoch": 0.02717867788883044,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.3536,
|
|
"mean_token_accuracy": 0.08135458827018738,
|
|
"num_tokens": 1008676.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.800386238098144,
|
|
"epoch": 0.027470921737097434,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.4651,
|
|
"mean_token_accuracy": 0.07976322658360005,
|
|
"num_tokens": 1019951.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.785122346878052,
|
|
"epoch": 0.027763165585364426,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.3981,
|
|
"mean_token_accuracy": 0.0761278249323368,
|
|
"num_tokens": 1030219.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.781033182144165,
|
|
"epoch": 0.028055409433631422,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.4141,
|
|
"mean_token_accuracy": 0.07795999571681023,
|
|
"num_tokens": 1041911.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.891617631912231,
|
|
"epoch": 0.028347653281898415,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.4312,
|
|
"mean_token_accuracy": 0.08173990175127983,
|
|
"num_tokens": 1052638.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.774001789093018,
|
|
"epoch": 0.02863989713016541,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.3755,
|
|
"mean_token_accuracy": 0.08225451484322548,
|
|
"num_tokens": 1063465.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.861321496963501,
|
|
"epoch": 0.028932140978432403,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.3888,
|
|
"mean_token_accuracy": 0.08350497037172318,
|
|
"num_tokens": 1073583.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.868850564956665,
|
|
"epoch": 0.0292243848266994,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.4377,
|
|
"mean_token_accuracy": 0.08040728494524955,
|
|
"num_tokens": 1083961.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.663293409347534,
|
|
"epoch": 0.02951662867496639,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.3053,
|
|
"mean_token_accuracy": 0.08032148070633412,
|
|
"num_tokens": 1095339.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.742640733718872,
|
|
"epoch": 0.029808872523233387,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.3746,
|
|
"mean_token_accuracy": 0.07911042794585228,
|
|
"num_tokens": 1105518.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.822371339797973,
|
|
"epoch": 0.03010111637150038,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.2683,
|
|
"mean_token_accuracy": 0.08846351876854897,
|
|
"num_tokens": 1115666.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.7058703899383545,
|
|
"epoch": 0.030393360219767372,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.3901,
|
|
"mean_token_accuracy": 0.08510988727211952,
|
|
"num_tokens": 1125827.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.762892246246338,
|
|
"epoch": 0.030685604068034368,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.3453,
|
|
"mean_token_accuracy": 0.08135025277733803,
|
|
"num_tokens": 1136154.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.712664318084717,
|
|
"epoch": 0.03097784791630136,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.3361,
|
|
"mean_token_accuracy": 0.08622926697134972,
|
|
"num_tokens": 1146392.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.7614504337310795,
|
|
"epoch": 0.03127009176456835,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.3515,
|
|
"mean_token_accuracy": 0.08140693604946136,
|
|
"num_tokens": 1157566.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.6518913269042965,
|
|
"epoch": 0.03156233561283535,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.3194,
|
|
"mean_token_accuracy": 0.0838658906519413,
|
|
"num_tokens": 1167795.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.775746774673462,
|
|
"epoch": 0.031854579461102345,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.2691,
|
|
"mean_token_accuracy": 0.09419989511370659,
|
|
"num_tokens": 1178894.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.634786605834961,
|
|
"epoch": 0.03214682330936934,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.2696,
|
|
"mean_token_accuracy": 0.08667804896831513,
|
|
"num_tokens": 1189361.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.752673006057739,
|
|
"epoch": 0.03243906715763633,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.3135,
|
|
"mean_token_accuracy": 0.08304268047213555,
|
|
"num_tokens": 1198992.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.692470407485962,
|
|
"epoch": 0.032731311005903325,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.294,
|
|
"mean_token_accuracy": 0.07943044230341911,
|
|
"num_tokens": 1208732.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.655223751068116,
|
|
"epoch": 0.03302355485417032,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.2954,
|
|
"mean_token_accuracy": 0.08259731158614159,
|
|
"num_tokens": 1219015.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.612499952316284,
|
|
"epoch": 0.03331579870243731,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.1795,
|
|
"mean_token_accuracy": 0.1000419594347477,
|
|
"num_tokens": 1229227.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.750643157958985,
|
|
"epoch": 0.033608042550704306,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.3098,
|
|
"mean_token_accuracy": 0.08021004945039749,
|
|
"num_tokens": 1239654.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.618366050720215,
|
|
"epoch": 0.0339002863989713,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.3005,
|
|
"mean_token_accuracy": 0.08035954535007477,
|
|
"num_tokens": 1250955.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.729614543914795,
|
|
"epoch": 0.0341925302472383,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.2497,
|
|
"mean_token_accuracy": 0.08898582234978676,
|
|
"num_tokens": 1262574.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.672766494750976,
|
|
"epoch": 0.03448477409550529,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.2059,
|
|
"mean_token_accuracy": 0.08972268775105477,
|
|
"num_tokens": 1273272.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.505560779571534,
|
|
"epoch": 0.03477701794377228,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.1898,
|
|
"mean_token_accuracy": 0.09178911671042442,
|
|
"num_tokens": 1284001.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.633696794509888,
|
|
"epoch": 0.03506926179203928,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.2354,
|
|
"mean_token_accuracy": 0.08889753520488738,
|
|
"num_tokens": 1294671.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.695655345916748,
|
|
"epoch": 0.035361505640306275,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.2349,
|
|
"mean_token_accuracy": 0.09624799937009812,
|
|
"num_tokens": 1305397.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.568394422531128,
|
|
"epoch": 0.035653749488573264,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.2377,
|
|
"mean_token_accuracy": 0.08420243933796882,
|
|
"num_tokens": 1317449.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.562969303131103,
|
|
"epoch": 0.03594599333684026,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.176,
|
|
"mean_token_accuracy": 0.08269712552428246,
|
|
"num_tokens": 1328645.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.559274625778198,
|
|
"epoch": 0.036238237185107255,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.1652,
|
|
"mean_token_accuracy": 0.09311013892292977,
|
|
"num_tokens": 1339815.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.583710241317749,
|
|
"epoch": 0.036530481033374244,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.1795,
|
|
"mean_token_accuracy": 0.08754727616906166,
|
|
"num_tokens": 1350301.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.535983610153198,
|
|
"epoch": 0.03682272488164124,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.1724,
|
|
"mean_token_accuracy": 0.08950133547186852,
|
|
"num_tokens": 1360928.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.625943231582641,
|
|
"epoch": 0.037114968729908236,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.1957,
|
|
"mean_token_accuracy": 0.08902273774147033,
|
|
"num_tokens": 1371618.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.511281824111938,
|
|
"epoch": 0.03740721257817523,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.1562,
|
|
"mean_token_accuracy": 0.08582355380058289,
|
|
"num_tokens": 1382824.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.470151281356811,
|
|
"epoch": 0.03769945642644222,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.1191,
|
|
"mean_token_accuracy": 0.09281812384724616,
|
|
"num_tokens": 1393244.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.475745964050293,
|
|
"epoch": 0.03799170027470922,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.0909,
|
|
"mean_token_accuracy": 0.09678038656711578,
|
|
"num_tokens": 1403469.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.615570402145385,
|
|
"epoch": 0.03828394412297621,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.1651,
|
|
"mean_token_accuracy": 0.09157615751028061,
|
|
"num_tokens": 1413860.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.553770732879639,
|
|
"epoch": 0.03857618797124321,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.2248,
|
|
"mean_token_accuracy": 0.08392854556441307,
|
|
"num_tokens": 1423985.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.531700229644775,
|
|
"epoch": 0.0388684318195102,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.1654,
|
|
"mean_token_accuracy": 0.09291392266750335,
|
|
"num_tokens": 1434838.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.482819223403931,
|
|
"epoch": 0.039160675667777194,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.0905,
|
|
"mean_token_accuracy": 0.09290852397680283,
|
|
"num_tokens": 1446429.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.611834764480591,
|
|
"epoch": 0.03945291951604419,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.1879,
|
|
"mean_token_accuracy": 0.08893333822488785,
|
|
"num_tokens": 1456726.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.5248387336730955,
|
|
"epoch": 0.03974516336431118,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.0781,
|
|
"mean_token_accuracy": 0.09758968204259873,
|
|
"num_tokens": 1466378.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.323547267913819,
|
|
"epoch": 0.040037407212578174,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.0345,
|
|
"mean_token_accuracy": 0.09719507321715355,
|
|
"num_tokens": 1477755.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.507241344451904,
|
|
"epoch": 0.04032965106084517,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.1006,
|
|
"mean_token_accuracy": 0.09118306115269662,
|
|
"num_tokens": 1488209.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.419371461868286,
|
|
"epoch": 0.040621894909112166,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.1465,
|
|
"mean_token_accuracy": 0.08755657598376274,
|
|
"num_tokens": 1498658.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.546753549575806,
|
|
"epoch": 0.040914138757379155,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.1576,
|
|
"mean_token_accuracy": 0.08506937101483344,
|
|
"num_tokens": 1509766.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.439605093002319,
|
|
"epoch": 0.04120638260564615,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.036,
|
|
"mean_token_accuracy": 0.0939096562564373,
|
|
"num_tokens": 1521139.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.4743701934814455,
|
|
"epoch": 0.04149862645391315,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.1115,
|
|
"mean_token_accuracy": 0.08748833164572715,
|
|
"num_tokens": 1531875.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.395281553268433,
|
|
"epoch": 0.041790870302180136,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.0801,
|
|
"mean_token_accuracy": 0.08955910429358482,
|
|
"num_tokens": 1542678.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.462686967849732,
|
|
"epoch": 0.04208311415044713,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.0268,
|
|
"mean_token_accuracy": 0.09358467385172844,
|
|
"num_tokens": 1553882.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.43300313949585,
|
|
"epoch": 0.04237535799871413,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.1176,
|
|
"mean_token_accuracy": 0.08562448695302009,
|
|
"num_tokens": 1564854.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.425636720657349,
|
|
"epoch": 0.042667601846981124,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.0026,
|
|
"mean_token_accuracy": 0.09751182347536087,
|
|
"num_tokens": 1576112.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.374439096450805,
|
|
"epoch": 0.04295984569524811,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.0378,
|
|
"mean_token_accuracy": 0.09424405097961426,
|
|
"num_tokens": 1586235.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.347483777999878,
|
|
"epoch": 0.04325208954351511,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.0347,
|
|
"mean_token_accuracy": 0.0911446213722229,
|
|
"num_tokens": 1596727.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.407256555557251,
|
|
"epoch": 0.043544333391782104,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.9573,
|
|
"mean_token_accuracy": 0.09358611106872558,
|
|
"num_tokens": 1607008.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.389186716079712,
|
|
"epoch": 0.0438365772400491,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.0579,
|
|
"mean_token_accuracy": 0.0928824670612812,
|
|
"num_tokens": 1618081.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.385285234451294,
|
|
"epoch": 0.04412882108831609,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.0165,
|
|
"mean_token_accuracy": 0.08930294588208199,
|
|
"num_tokens": 1629095.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.394097995758057,
|
|
"epoch": 0.044421064936583085,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.0118,
|
|
"mean_token_accuracy": 0.093733761459589,
|
|
"num_tokens": 1639861.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.372588682174682,
|
|
"epoch": 0.04471330878485008,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.0534,
|
|
"mean_token_accuracy": 0.09636619538068772,
|
|
"num_tokens": 1649169.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.278450775146484,
|
|
"epoch": 0.04500555263311707,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.9357,
|
|
"mean_token_accuracy": 0.1010350152850151,
|
|
"num_tokens": 1658528.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.303726387023926,
|
|
"epoch": 0.045297796481384066,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.0068,
|
|
"mean_token_accuracy": 0.09450188651680946,
|
|
"num_tokens": 1669231.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.309107494354248,
|
|
"epoch": 0.04559004032965106,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.9475,
|
|
"mean_token_accuracy": 0.09322728961706161,
|
|
"num_tokens": 1680503.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.336097192764282,
|
|
"epoch": 0.04588228417791806,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.9766,
|
|
"mean_token_accuracy": 0.09063231274485588,
|
|
"num_tokens": 1690827.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.370522546768188,
|
|
"epoch": 0.04617452802618505,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.0418,
|
|
"mean_token_accuracy": 0.09461386948823929,
|
|
"num_tokens": 1702931.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.289389944076538,
|
|
"epoch": 0.04646677187445204,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.9307,
|
|
"mean_token_accuracy": 0.09922353029251099,
|
|
"num_tokens": 1712834.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.3177947998046875,
|
|
"epoch": 0.04675901572271904,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.0006,
|
|
"mean_token_accuracy": 0.0919931598007679,
|
|
"num_tokens": 1723289.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.245712804794311,
|
|
"epoch": 0.04705125957098603,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.9702,
|
|
"mean_token_accuracy": 0.09567693397402763,
|
|
"num_tokens": 1734477.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.306273651123047,
|
|
"epoch": 0.04734350341925302,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.9762,
|
|
"mean_token_accuracy": 0.09110540002584458,
|
|
"num_tokens": 1745187.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.276857900619507,
|
|
"epoch": 0.04763574726752002,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.9011,
|
|
"mean_token_accuracy": 0.10115249156951904,
|
|
"num_tokens": 1756687.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.304260778427124,
|
|
"epoch": 0.047927991115787015,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.024,
|
|
"mean_token_accuracy": 0.09978155121207237,
|
|
"num_tokens": 1767928.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.309319734573364,
|
|
"epoch": 0.048220234964054004,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.9585,
|
|
"mean_token_accuracy": 0.09580064937472343,
|
|
"num_tokens": 1777951.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.194837522506714,
|
|
"epoch": 0.048512478812321,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.9514,
|
|
"mean_token_accuracy": 0.09750414118170739,
|
|
"num_tokens": 1788598.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.25841178894043,
|
|
"epoch": 0.048804722660587996,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.9107,
|
|
"mean_token_accuracy": 0.10313281938433647,
|
|
"num_tokens": 1799446.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.308444929122925,
|
|
"epoch": 0.04909696650885499,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.93,
|
|
"mean_token_accuracy": 0.09639307036995888,
|
|
"num_tokens": 1809687.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.247332191467285,
|
|
"epoch": 0.04938921035712198,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.9496,
|
|
"mean_token_accuracy": 0.10043755695223808,
|
|
"num_tokens": 1820535.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.192328262329101,
|
|
"epoch": 0.04968145420538898,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.8901,
|
|
"mean_token_accuracy": 0.09903936088085175,
|
|
"num_tokens": 1830834.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.244427633285523,
|
|
"epoch": 0.04997369805365597,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.8568,
|
|
"mean_token_accuracy": 0.10095973312854767,
|
|
"num_tokens": 1842615.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.1798933982849125,
|
|
"epoch": 0.05026594190192296,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.8778,
|
|
"mean_token_accuracy": 0.10051687061786652,
|
|
"num_tokens": 1852844.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.210625696182251,
|
|
"epoch": 0.05055818575018996,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.9139,
|
|
"mean_token_accuracy": 0.09776052087545395,
|
|
"num_tokens": 1862528.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.177963972091675,
|
|
"epoch": 0.05085042959845695,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.8757,
|
|
"mean_token_accuracy": 0.1031526930630207,
|
|
"num_tokens": 1873275.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.275309658050537,
|
|
"epoch": 0.05114267344672395,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.9153,
|
|
"mean_token_accuracy": 0.10134659111499786,
|
|
"num_tokens": 1884724.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.248415899276734,
|
|
"epoch": 0.05143491729499094,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.8861,
|
|
"mean_token_accuracy": 0.09837430268526078,
|
|
"num_tokens": 1894705.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.133459377288818,
|
|
"epoch": 0.051727161143257934,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.9485,
|
|
"mean_token_accuracy": 0.09655695110559463,
|
|
"num_tokens": 1906553.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.257501792907715,
|
|
"epoch": 0.05201940499152493,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.8849,
|
|
"mean_token_accuracy": 0.1023280717432499,
|
|
"num_tokens": 1916628.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.0709809303283695,
|
|
"epoch": 0.052311648839791926,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.725,
|
|
"mean_token_accuracy": 0.11589453369379044,
|
|
"num_tokens": 1926686.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.150864934921264,
|
|
"epoch": 0.052603892688058915,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.9068,
|
|
"mean_token_accuracy": 0.101015555113554,
|
|
"num_tokens": 1938150.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.1812304019927975,
|
|
"epoch": 0.05289613653632591,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.8918,
|
|
"mean_token_accuracy": 0.097091793268919,
|
|
"num_tokens": 1948617.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.160140323638916,
|
|
"epoch": 0.053188380384592907,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.8934,
|
|
"mean_token_accuracy": 0.10218235999345779,
|
|
"num_tokens": 1958943.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.165960168838501,
|
|
"epoch": 0.053480624232859895,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.8192,
|
|
"mean_token_accuracy": 0.10898060277104378,
|
|
"num_tokens": 1968988.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.051486253738403,
|
|
"epoch": 0.05377286808112689,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.7949,
|
|
"mean_token_accuracy": 0.10556035414338112,
|
|
"num_tokens": 1979490.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.28261432647705,
|
|
"epoch": 0.05406511192939389,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.8574,
|
|
"mean_token_accuracy": 0.10469735637307168,
|
|
"num_tokens": 1990377.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.070431137084961,
|
|
"epoch": 0.05435735577766088,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.8361,
|
|
"mean_token_accuracy": 0.10013288781046867,
|
|
"num_tokens": 2001678.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.103927278518677,
|
|
"epoch": 0.05464959962592787,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.8266,
|
|
"mean_token_accuracy": 0.10559705421328544,
|
|
"num_tokens": 2011550.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.146097993850708,
|
|
"epoch": 0.05494184347419487,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.8591,
|
|
"mean_token_accuracy": 0.10166500806808472,
|
|
"num_tokens": 2021607.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.074561977386475,
|
|
"epoch": 0.055234087322461864,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.831,
|
|
"mean_token_accuracy": 0.10222307071089745,
|
|
"num_tokens": 2030975.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.097246503829956,
|
|
"epoch": 0.05552633117072885,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.9034,
|
|
"mean_token_accuracy": 0.09934694394469261,
|
|
"num_tokens": 2041799.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.116580200195313,
|
|
"epoch": 0.05581857501899585,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.8168,
|
|
"mean_token_accuracy": 0.10345752537250519,
|
|
"num_tokens": 2052960.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.11494288444519,
|
|
"epoch": 0.056110818867262845,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.8662,
|
|
"mean_token_accuracy": 0.09989891946315765,
|
|
"num_tokens": 2064156.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.182250165939331,
|
|
"epoch": 0.05640306271552984,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.8254,
|
|
"mean_token_accuracy": 0.10051817372441292,
|
|
"num_tokens": 2074774.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.077956199645996,
|
|
"epoch": 0.05669530656379683,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.8527,
|
|
"mean_token_accuracy": 0.10035398751497268,
|
|
"num_tokens": 2086164.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.885358047485352,
|
|
"epoch": 0.056987550412063825,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.672,
|
|
"mean_token_accuracy": 0.127625822275877,
|
|
"num_tokens": 2096268.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.233099365234375,
|
|
"epoch": 0.05727979426033082,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.8813,
|
|
"mean_token_accuracy": 0.09680663272738457,
|
|
"num_tokens": 2108511.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.136028242111206,
|
|
"epoch": 0.05757203810859782,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.8085,
|
|
"mean_token_accuracy": 0.0988966628909111,
|
|
"num_tokens": 2119622.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.998387289047241,
|
|
"epoch": 0.057864281956864806,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.8613,
|
|
"mean_token_accuracy": 0.10771243944764138,
|
|
"num_tokens": 2131021.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.106973361968994,
|
|
"epoch": 0.0581565258051318,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.8496,
|
|
"mean_token_accuracy": 0.10358692184090615,
|
|
"num_tokens": 2141797.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.108656787872315,
|
|
"epoch": 0.0584487696533988,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.7338,
|
|
"mean_token_accuracy": 0.10600685104727745,
|
|
"num_tokens": 2153609.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.041635751724243,
|
|
"epoch": 0.05874101350166579,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.8544,
|
|
"mean_token_accuracy": 0.10038745403289795,
|
|
"num_tokens": 2164237.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.159450626373291,
|
|
"epoch": 0.05903325734993278,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.8092,
|
|
"mean_token_accuracy": 0.09949537962675095,
|
|
"num_tokens": 2175035.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.941095781326294,
|
|
"epoch": 0.05932550119819978,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.7273,
|
|
"mean_token_accuracy": 0.10634237006306649,
|
|
"num_tokens": 2186217.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.0672167301177975,
|
|
"epoch": 0.059617745046466775,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.777,
|
|
"mean_token_accuracy": 0.10704884603619576,
|
|
"num_tokens": 2197290.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.999376821517944,
|
|
"epoch": 0.059909988894733764,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.8155,
|
|
"mean_token_accuracy": 0.0990687295794487,
|
|
"num_tokens": 2208835.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.138485765457153,
|
|
"epoch": 0.06020223274300076,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.7249,
|
|
"mean_token_accuracy": 0.10986540690064431,
|
|
"num_tokens": 2219153.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.9783261775970455,
|
|
"epoch": 0.060494476591267755,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.7549,
|
|
"mean_token_accuracy": 0.10616105198860168,
|
|
"num_tokens": 2230289.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.989655160903931,
|
|
"epoch": 0.060786720439534744,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.8395,
|
|
"mean_token_accuracy": 0.10555440336465835,
|
|
"num_tokens": 2241768.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.044928169250488,
|
|
"epoch": 0.06107896428780174,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.7114,
|
|
"mean_token_accuracy": 0.11182207837700844,
|
|
"num_tokens": 2253249.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.989438867568969,
|
|
"epoch": 0.061371208136068736,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.7792,
|
|
"mean_token_accuracy": 0.09858244881033898,
|
|
"num_tokens": 2263961.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.941675472259521,
|
|
"epoch": 0.06166345198433573,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.6945,
|
|
"mean_token_accuracy": 0.11112689971923828,
|
|
"num_tokens": 2274016.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.96781735420227,
|
|
"epoch": 0.06195569583260272,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.7391,
|
|
"mean_token_accuracy": 0.10291829779744148,
|
|
"num_tokens": 2285164.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.105303335189819,
|
|
"epoch": 0.06224793968086972,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.7238,
|
|
"mean_token_accuracy": 0.10441325902938843,
|
|
"num_tokens": 2294887.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.961658096313476,
|
|
"epoch": 0.0625401835291367,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.7424,
|
|
"mean_token_accuracy": 0.10544760972261429,
|
|
"num_tokens": 2305787.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.00815224647522,
|
|
"epoch": 0.0628324273774037,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.7683,
|
|
"mean_token_accuracy": 0.10668534934520721,
|
|
"num_tokens": 2316756.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.019680690765381,
|
|
"epoch": 0.0631246712256707,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.783,
|
|
"mean_token_accuracy": 0.10868853554129601,
|
|
"num_tokens": 2327335.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.018626260757446,
|
|
"epoch": 0.0634169150739377,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.7444,
|
|
"mean_token_accuracy": 0.1080512523651123,
|
|
"num_tokens": 2337182.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.036516618728638,
|
|
"epoch": 0.06370915892220469,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.7131,
|
|
"mean_token_accuracy": 0.10611762776970864,
|
|
"num_tokens": 2347843.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.903224134445191,
|
|
"epoch": 0.06400140277047169,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.7136,
|
|
"mean_token_accuracy": 0.10907531604170799,
|
|
"num_tokens": 2358159.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.897138977050782,
|
|
"epoch": 0.06429364661873868,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.6937,
|
|
"mean_token_accuracy": 0.10973558053374291,
|
|
"num_tokens": 2368191.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.996454906463623,
|
|
"epoch": 0.06458589046700566,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.6386,
|
|
"mean_token_accuracy": 0.11236895397305488,
|
|
"num_tokens": 2377672.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.894516134262085,
|
|
"epoch": 0.06487813431527266,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.7148,
|
|
"mean_token_accuracy": 0.11475707441568375,
|
|
"num_tokens": 2387612.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.9777552604675295,
|
|
"epoch": 0.06517037816353966,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.6564,
|
|
"mean_token_accuracy": 0.11448713690042496,
|
|
"num_tokens": 2397198.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.946916341781616,
|
|
"epoch": 0.06546262201180665,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.6935,
|
|
"mean_token_accuracy": 0.11213591918349267,
|
|
"num_tokens": 2407816.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.891662502288819,
|
|
"epoch": 0.06575486586007365,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.6816,
|
|
"mean_token_accuracy": 0.10039346590638161,
|
|
"num_tokens": 2419537.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.98465371131897,
|
|
"epoch": 0.06604710970834064,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.6259,
|
|
"mean_token_accuracy": 0.10811248049139977,
|
|
"num_tokens": 2429043.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.832172632217407,
|
|
"epoch": 0.06633935355660764,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.6566,
|
|
"mean_token_accuracy": 0.11272674351930619,
|
|
"num_tokens": 2439469.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.94088830947876,
|
|
"epoch": 0.06663159740487462,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.6637,
|
|
"mean_token_accuracy": 0.1104379877448082,
|
|
"num_tokens": 2449831.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.871842384338379,
|
|
"epoch": 0.06692384125314162,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.6029,
|
|
"mean_token_accuracy": 0.1185630053281784,
|
|
"num_tokens": 2460709.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.824612331390381,
|
|
"epoch": 0.06721608510140861,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.6444,
|
|
"mean_token_accuracy": 0.11218708977103234,
|
|
"num_tokens": 2472246.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.884087419509887,
|
|
"epoch": 0.06750832894967561,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.6262,
|
|
"mean_token_accuracy": 0.11107529774308204,
|
|
"num_tokens": 2482793.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.801684379577637,
|
|
"epoch": 0.0678005727979426,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.6194,
|
|
"mean_token_accuracy": 0.117616818100214,
|
|
"num_tokens": 2493186.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.965762662887573,
|
|
"epoch": 0.0680928166462096,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.6642,
|
|
"mean_token_accuracy": 0.1107054591178894,
|
|
"num_tokens": 2505100.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.8852849960327145,
|
|
"epoch": 0.0683850604944766,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.6706,
|
|
"mean_token_accuracy": 0.10744462683796882,
|
|
"num_tokens": 2515373.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.896916151046753,
|
|
"epoch": 0.06867730434274359,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.6578,
|
|
"mean_token_accuracy": 0.11280061826109886,
|
|
"num_tokens": 2526645.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.876949310302734,
|
|
"epoch": 0.06896954819101057,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.6726,
|
|
"mean_token_accuracy": 0.10913796946406365,
|
|
"num_tokens": 2537109.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.938536024093628,
|
|
"epoch": 0.06926179203927757,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.6137,
|
|
"mean_token_accuracy": 0.11598925217986107,
|
|
"num_tokens": 2548709.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.8740095615386965,
|
|
"epoch": 0.06955403588754457,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.6629,
|
|
"mean_token_accuracy": 0.1084805577993393,
|
|
"num_tokens": 2559802.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.837321758270264,
|
|
"epoch": 0.06984627973581156,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.6605,
|
|
"mean_token_accuracy": 0.10946222245693207,
|
|
"num_tokens": 2570343.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.8656223773956295,
|
|
"epoch": 0.07013852358407856,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.6447,
|
|
"mean_token_accuracy": 0.10875394120812416,
|
|
"num_tokens": 2580772.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.828289127349853,
|
|
"epoch": 0.07043076743234555,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.6375,
|
|
"mean_token_accuracy": 0.11186778992414474,
|
|
"num_tokens": 2592383.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.913778018951416,
|
|
"epoch": 0.07072301128061255,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.6364,
|
|
"mean_token_accuracy": 0.10910762697458268,
|
|
"num_tokens": 2602807.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.835033416748047,
|
|
"epoch": 0.07101525512887953,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.629,
|
|
"mean_token_accuracy": 0.11060894504189492,
|
|
"num_tokens": 2613978.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.906363582611084,
|
|
"epoch": 0.07130749897714653,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.6075,
|
|
"mean_token_accuracy": 0.11025396510958671,
|
|
"num_tokens": 2624044.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.882346868515015,
|
|
"epoch": 0.07159974282541352,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.6085,
|
|
"mean_token_accuracy": 0.11288385465741158,
|
|
"num_tokens": 2635550.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.821655368804931,
|
|
"epoch": 0.07189198667368052,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.5774,
|
|
"mean_token_accuracy": 0.11880745887756347,
|
|
"num_tokens": 2645343.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.8402406692504885,
|
|
"epoch": 0.07218423052194751,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.6189,
|
|
"mean_token_accuracy": 0.11299417838454247,
|
|
"num_tokens": 2655366.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.900226879119873,
|
|
"epoch": 0.07247647437021451,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.6044,
|
|
"mean_token_accuracy": 0.11406284347176551,
|
|
"num_tokens": 2666438.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.799708652496338,
|
|
"epoch": 0.0727687182184815,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.5675,
|
|
"mean_token_accuracy": 0.1110381007194519,
|
|
"num_tokens": 2677800.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.799835634231568,
|
|
"epoch": 0.07306096206674849,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.5575,
|
|
"mean_token_accuracy": 0.1164891928434372,
|
|
"num_tokens": 2688876.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.823706436157226,
|
|
"epoch": 0.07335320591501548,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.5231,
|
|
"mean_token_accuracy": 0.12699810415506363,
|
|
"num_tokens": 2700265.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.76237301826477,
|
|
"epoch": 0.07364544976328248,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.555,
|
|
"mean_token_accuracy": 0.12002863585948945,
|
|
"num_tokens": 2711336.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.826159906387329,
|
|
"epoch": 0.07393769361154948,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.6281,
|
|
"mean_token_accuracy": 0.11225389763712883,
|
|
"num_tokens": 2722138.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.837359285354614,
|
|
"epoch": 0.07422993745981647,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.5622,
|
|
"mean_token_accuracy": 0.11845961436629296,
|
|
"num_tokens": 2733016.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.825304698944092,
|
|
"epoch": 0.07452218130808347,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.5785,
|
|
"mean_token_accuracy": 0.11900259032845498,
|
|
"num_tokens": 2743424.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.865557479858398,
|
|
"epoch": 0.07481442515635046,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.616,
|
|
"mean_token_accuracy": 0.11359821557998658,
|
|
"num_tokens": 2753867.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.791312026977539,
|
|
"epoch": 0.07510666900461745,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.5132,
|
|
"mean_token_accuracy": 0.11576777994632721,
|
|
"num_tokens": 2764602.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.750291347503662,
|
|
"epoch": 0.07539891285288444,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.627,
|
|
"mean_token_accuracy": 0.11037226766347885,
|
|
"num_tokens": 2775907.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.848241567611694,
|
|
"epoch": 0.07569115670115144,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.5906,
|
|
"mean_token_accuracy": 0.11550377830862998,
|
|
"num_tokens": 2787455.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.756697750091552,
|
|
"epoch": 0.07598340054941843,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.5252,
|
|
"mean_token_accuracy": 0.12691028639674187,
|
|
"num_tokens": 2797600.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.748210525512695,
|
|
"epoch": 0.07627564439768543,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.4557,
|
|
"mean_token_accuracy": 0.12226552590727806,
|
|
"num_tokens": 2808056.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.778938245773316,
|
|
"epoch": 0.07656788824595243,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.511,
|
|
"mean_token_accuracy": 0.11900310069322587,
|
|
"num_tokens": 2817962.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.7642310619354244,
|
|
"epoch": 0.07686013209421942,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.5278,
|
|
"mean_token_accuracy": 0.12055002972483635,
|
|
"num_tokens": 2828769.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.761574554443359,
|
|
"epoch": 0.07715237594248642,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.4627,
|
|
"mean_token_accuracy": 0.12216627746820449,
|
|
"num_tokens": 2838983.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.7704860210418705,
|
|
"epoch": 0.0774446197907534,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.5237,
|
|
"mean_token_accuracy": 0.10742606818675995,
|
|
"num_tokens": 2848892.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.73878755569458,
|
|
"epoch": 0.0777368636390204,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.471,
|
|
"mean_token_accuracy": 0.12039833962917328,
|
|
"num_tokens": 2859303.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.816798734664917,
|
|
"epoch": 0.07802910748728739,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.696,
|
|
"mean_token_accuracy": 0.10775517970323563,
|
|
"num_tokens": 2871318.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.841889810562134,
|
|
"epoch": 0.07832135133555439,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.516,
|
|
"mean_token_accuracy": 0.11806261539459229,
|
|
"num_tokens": 2882150.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.674707317352295,
|
|
"epoch": 0.07861359518382138,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.4974,
|
|
"mean_token_accuracy": 0.1205772653222084,
|
|
"num_tokens": 2892657.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.792233276367187,
|
|
"epoch": 0.07890583903208838,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.4611,
|
|
"mean_token_accuracy": 0.12206808105111122,
|
|
"num_tokens": 2903298.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.646578741073609,
|
|
"epoch": 0.07919808288035537,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.5163,
|
|
"mean_token_accuracy": 0.11933484375476837,
|
|
"num_tokens": 2914684.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.808360195159912,
|
|
"epoch": 0.07949032672862236,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.4423,
|
|
"mean_token_accuracy": 0.12348946481943131,
|
|
"num_tokens": 2925191.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.701859855651856,
|
|
"epoch": 0.07978257057688935,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.5226,
|
|
"mean_token_accuracy": 0.12045117989182472,
|
|
"num_tokens": 2935498.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.793068552017212,
|
|
"epoch": 0.08007481442515635,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.4408,
|
|
"mean_token_accuracy": 0.12177050709724427,
|
|
"num_tokens": 2945590.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.6989601135253904,
|
|
"epoch": 0.08036705827342334,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.415,
|
|
"mean_token_accuracy": 0.1295503109693527,
|
|
"num_tokens": 2954959.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.739167213439941,
|
|
"epoch": 0.08065930212169034,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.4962,
|
|
"mean_token_accuracy": 0.12157377004623413,
|
|
"num_tokens": 2964808.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.74145622253418,
|
|
"epoch": 0.08095154596995734,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.4832,
|
|
"mean_token_accuracy": 0.12358191087841988,
|
|
"num_tokens": 2975032.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.724349927902222,
|
|
"epoch": 0.08124378981822433,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.4382,
|
|
"mean_token_accuracy": 0.12445014715194702,
|
|
"num_tokens": 2984426.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.618544483184815,
|
|
"epoch": 0.08153603366649131,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.5059,
|
|
"mean_token_accuracy": 0.11675598993897437,
|
|
"num_tokens": 2994871.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.759203100204468,
|
|
"epoch": 0.08182827751475831,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.4283,
|
|
"mean_token_accuracy": 0.1238317184150219,
|
|
"num_tokens": 3004800.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.709669494628907,
|
|
"epoch": 0.0821205213630253,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.5211,
|
|
"mean_token_accuracy": 0.11378786638379097,
|
|
"num_tokens": 3015561.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.706784200668335,
|
|
"epoch": 0.0824127652112923,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.4702,
|
|
"mean_token_accuracy": 0.11403566598892212,
|
|
"num_tokens": 3025885.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.691977500915527,
|
|
"epoch": 0.0827050090595593,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.4026,
|
|
"mean_token_accuracy": 0.12606138437986375,
|
|
"num_tokens": 3035524.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.703480625152588,
|
|
"epoch": 0.0829972529078263,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.5224,
|
|
"mean_token_accuracy": 0.11795256957411766,
|
|
"num_tokens": 3046957.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.78633394241333,
|
|
"epoch": 0.08328949675609329,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.4441,
|
|
"mean_token_accuracy": 0.11978953778743744,
|
|
"num_tokens": 3056713.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.641093826293945,
|
|
"epoch": 0.08358174060436027,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.432,
|
|
"mean_token_accuracy": 0.12080147936940193,
|
|
"num_tokens": 3068369.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.655155420303345,
|
|
"epoch": 0.08387398445262727,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.4762,
|
|
"mean_token_accuracy": 0.12348118275403977,
|
|
"num_tokens": 3079083.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.710379791259766,
|
|
"epoch": 0.08416622830089426,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.5061,
|
|
"mean_token_accuracy": 0.11596815660595894,
|
|
"num_tokens": 3089150.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.672757816314697,
|
|
"epoch": 0.08445847214916126,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.416,
|
|
"mean_token_accuracy": 0.12834767252206802,
|
|
"num_tokens": 3098696.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.736963558197021,
|
|
"epoch": 0.08475071599742826,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.4355,
|
|
"mean_token_accuracy": 0.120185237377882,
|
|
"num_tokens": 3108879.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.7161966323852536,
|
|
"epoch": 0.08504295984569525,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.3953,
|
|
"mean_token_accuracy": 0.12836327031254768,
|
|
"num_tokens": 3118756.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.680994129180908,
|
|
"epoch": 0.08533520369396225,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.4487,
|
|
"mean_token_accuracy": 0.1232797771692276,
|
|
"num_tokens": 3129130.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.678233671188354,
|
|
"epoch": 0.08562744754222923,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.4116,
|
|
"mean_token_accuracy": 0.12097690925002098,
|
|
"num_tokens": 3139528.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.644343137741089,
|
|
"epoch": 0.08591969139049623,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.4864,
|
|
"mean_token_accuracy": 0.12138932570815086,
|
|
"num_tokens": 3149456.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.659511518478394,
|
|
"epoch": 0.08621193523876322,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.4144,
|
|
"mean_token_accuracy": 0.12389760985970497,
|
|
"num_tokens": 3159851.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.634909582138062,
|
|
"epoch": 0.08650417908703022,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.3856,
|
|
"mean_token_accuracy": 0.13249688521027564,
|
|
"num_tokens": 3169897.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.706868124008179,
|
|
"epoch": 0.08679642293529721,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.4994,
|
|
"mean_token_accuracy": 0.122072122246027,
|
|
"num_tokens": 3180696.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.659936094284058,
|
|
"epoch": 0.08708866678356421,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.4162,
|
|
"mean_token_accuracy": 0.12308593988418579,
|
|
"num_tokens": 3191836.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.62534909248352,
|
|
"epoch": 0.0873809106318312,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.3726,
|
|
"mean_token_accuracy": 0.12648854553699493,
|
|
"num_tokens": 3202009.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.55533013343811,
|
|
"epoch": 0.0876731544800982,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.3333,
|
|
"mean_token_accuracy": 0.13207703083753586,
|
|
"num_tokens": 3212707.0,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 707746595143680.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|